“Code No: 156BN E D JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDE B, Tech II Year I] Semester Examinations, August/September - MACHINE LEARNING | >, (Computer Scienceand Engineering) i ax. Marks: 75 e: 3 Hours ae Answer any five questions All questions earry equal marks 55 U9 Define Well-Posed problem. Illustrate any four examples, for Well-Posed pe ral A) Whit dé you mean by Candidate elimination? Explain; ff i) _ What are the concepts of learning as search? Discuss. Be Discuss the appropriate problems for decision tree learning. (8+7] CS Contrast the hypothesi$ space search in ID3 and candidate elimination algorithm. ~b) Explain the Back propagation learning algorithm and its limitations. (748) 4.2) Howa multi layered network leams usinig a gradient descent algorithmi?:Discuss. £ b) Explain the methods for comparing the accuracy of two hypotheses. [8+7] : State Bayes theorem. Illustrate Bayes theorem with an example. b) Describe the mistake bound model of learning, [8+7] 6.2) _ Explain Gibs algorithm with an example. i i _ ») State and explain the Minimum Description Length Principle, [847] 8 Discuss about Hypothesis space search in genetic algorithms. ) Write the basic algorithm for learning sets of First-Order Rules. [8+7] 8.) Discuss Explanation-Based learning of search control knowledge. 5) Explain the inductive analytical approaches to learning, —_ [8+7] ~--00000---